I use CHARMm forcefield to run MD for carbohydrate, and noticed that recently, when using existing and MMFF94 methods for partical charge assignment, the resulting CHARMM energy for the same molecule is about -25 and -300 kcal/mol, respectively. The total charge showed in CHARMM log file are correct for both methods (say, -2). However, when I click label for dsv file and choose "molecule" and add " partical charge" and "formal charge", I only see symbol "0" for both methods.  Does this mean there is a bug ?  Which method I should trust?